Pat Hanrahan on the Value of Visualization

Jennifer Vaughan, Product Consultant at Tableau, posts this about Pat Hanrahan's talk at Tableau Customer Conference 2011. Her notes: According to Pat Hanrahan, the value of visualization is to improve decision making, to be dynamic, and to facilitate coordination. There is no single best visualization and designing can be difficult, but think of visualizations as a symbolic form that enables problem solving. Taking a visualization one step further is to enable collaboration, for others to learn from you and make decisions. As a result, people and groups are more productive.

Jennifer Vaughan, Product Consultant at Tableau, posts this about Pat Hanrahan's talk at Tableau Customer Conference 2011. Her notes:

According to Pat Hanrahan, the value of visualization is to improve decision making, to be dynamic, and to facilitate coordination. There is no single best visualization and designing can be difficult, but think of visualizations as a symbolic form that enables problem solving. Taking a visualization one step further is to enable collaboration, for others to learn from you and make decisions. As a result, people and groups are more productive.

Visualizations are mathematical, symbolic and manipulable. They become more powerful at that point, since we can use them instead of just look at them. He calls this writing, when you can write in the visualization and not just read it. Pat helped invent the term visual analytics which means analytical processing and human reasoning facilitated by the interactive visualization. Essentially coupling the computing power, of algorithms that are highly interactive and visual, with people.

How has that been put into play within Tableau? VizSQL is the language. No wizards. Flexibility. And, the user can make formulas (or questions) for Tableau to create the visualization.

Take a table of information with headers. The headers are intelligent to create a view for the eyes to scan only what it wants to see, but when it comes down to scanning the rows for say a maximum value the brain is maxed at a couple pieces of information. Moving that same information into the bar chart the visualization is able to call attention that maximum value and allow the brain to focus on the data that is important. This is the principle behind the use of visualizations in Tableau. A lot of research has been conducted around this. An interesting fact, your pupils dilate when start to think, so using a tool like Tobii Technology, studies can be done on how much time is wasted on a confusing visualizations.

I will leave you with Pat's three important things about a visualization: reduce search time, reduce memory load, and allow perceptual interference. These are just a few things, but they are biggest things to think about.